| Low-skilled | High-skilled | |||
| Women (1) | Men (2) | Women (3) | Men (4) | |
| Panel A: Total Economy | ||||
| Dependent variable: log employment rate | ||||
| TP | 0.473* | 0.701*** | 0.054 | -0.034 |
| (0.284) | (0.136) | (0.101) | (0.054) | |
| N | 988 | 988 | 988 | 988 |
| R-squared | 0.85 | 0.90 | 0.40 | 0.85 |
| Panel B: Tradable and Nontradable Sector | ||||
| Dependent variable: log tradable sector employment rate | ||||
| TP | 2.679*** | 1.469*** | 1.009 | 0.998* |
| (0.712) | (0.349) | (0.853) | (0.523) | |
| N | 987 | 988 | 943 | 984 |
| R-squared | 0.67 | 0.92 | 0.56 | 0.75 |
| Dependent variable: log nontradable sector employment rate | ||||
| TP | -0.001 | -0.388** | -0.162 | 0.012 |
| (0.221) | (0.176) | (0.121) | (0.165) | |
| N | 988 | 988 | 988 | 988 |
| R-squared | 0.89 | 0.76 | 0.48 | 0.34 |
| State trends | Yes | Yes | Yes | Yes |
Notes: Microregion fixed-effects estimations. All dependent variables are measured as the log share of the relevant population, age 25-55. All estimations control for year-fixed effects, initial broad employment structure, and state-specific trends. Standard errors clustered by microregion are in parentheses, regressions are weighted by microregions’ 1991 population. * Significant at 10 percent; ** significant at 5 percent; *** significant at 1 percent.